SUMMARY OF WORK During the previous project period we adapted our algorithm to simulate stochastic local control of cardiac excitation-contraction coupling to run in parallel on clusters of computers linked by the message-passing interfact (MPI). Using the added computational power, we carried out simulations of the effects of buffers on calcium induced calcium release in both cardiac and skeletal muscle, assisting in the interpretation of controversial experimental results of studies that have tried to prove (or disprove) the role of local calcium-induced calcium release in skeletal muscle using high concentrations of intracellular BAPTA. We have also used this system to begin simulating the dynamics of calcium sparks, in order to interpret recent experimental results from our group that showed that sparks can be decomposed into smaller "quantal" units. The simulations showed that the unexpected fact that spark duration is inversely correlated with number of participating ryanodine receptors could, in fact, be explained by a CICR process with strong calcium-induced inactivation of ryanodine receptors. Future work will refine this model in correlation with experimental results, and provide a framework for interpreting studies on the role of L-type channel kinetics in determining EC coupling efficiency. We have also begun studies to model calcium-induced calcium release in pacemaker cells and primitive myocytes differentiated from embryonic stem cells. These simulations will help to understand the experimental results that show that CICR plays a critical role in the regulation of beating rate in these tissues. In order to study this phenomenon, our CICR model will be combined with differential-equation models of the cellular electrophysiology of the action potential in pacemaking cells.